Research: Caution on the equity risk premium
In this second article in a series on a new study, Nick Lyster and Amin Rajan argue that artificially low interest rates have dragged the predictive power of the equity risk premium to near zero
The equity risk premium (ERP), defined as the expected return of stocks in excess of a risk-free rate, has long been a weather vane of investors’ hopes and fears.
Outwardly, its recent high level in all markets might seem surprising, given the extent to which equity prices have recovered from their freefall in the global financial crisis in 2008.
Beneath the surface, however, the reason is simple; the ERP is high because bond yields are ultra-low. And yields are low partly as a result of the large-scale bond purchases under the quantitative easing (QE) programmes in the US, Europe and Japan, and partly due to the worries about the state of the global economy, according to the 2015 The Principal/CREATE-Research Survey*.
Thus, when asked what may happen over the next three years, our respondents highlighted various possibilities for the ERP, some of which overlap (see figure). Behind them lie two salient points.
The main reason the ERP will remain volatile is because much could still go wrong with the Federal Reserve’s exit strategy. For example, if it postpones rate increases, inflationary pressures could build up, requiring even bigger reductions further down the line. In the mean time, markets may go into speculative overdrive, raising the spectre of violent corrections.
On the other hand, even if the Fed is able to normalise rates in an orderly way, recovery may stall owing to the increased drag from the sheer size of the debt mountain in the US – it has grown from $142trn (€130trn) in 2008 to an all-time high of $199trn in 2014.
The Fed’s dilemma is exacerbated by the uncertain outlook of the global economy. Indeed, there are worries that in Europe and Japan, growth might disappoint, as their QE programmes may well generate far smaller wealth effects.
Besides, for all the talk about reforms in Europe, Japan and the emerging markets, implementation remains painfully slow. Impacts remain out of sight, while hopes run ahead of expectations. That leaves unconventional monetary policy as the only tool to drive global growth.
Viewpoint: a French asset manager
“In developed markets, the ERP is at a 20-year high. Yet, its ability to predict future equity prices is zero. This is because the old notion of ‘risk-free’ assets has been losing its validity for a long time. The QE programmes merely hastened it.
“The ERP is defined as the additional rate of return that investors require to compensate them for the risk of holding stocks instead of a risk-free asset, typically the US 10-year Treasury Bond or six-month Treasury bill.
“By definition, such an asset is meant to exhibit three features: the lowest returns in the investment pecking order; no correlation with risky assets; and zero difference between expected and actual returns.
“In practice, from 1999 to 2009, the US 10-year Treasury bonds not only outperformed risky assets such as equities; their actual returns were also well above the expected ones. In 2014, the 10-year German bund returned 40%.
“In fact, government bonds have violated every tenet of conventional investment wisdom over the past 30 years. At 14.5% in 2013, the ERP in the US was at its highest in 50 years and well above 10.5% during the financial crisis in 2009.
“Yet, it tells us little about how equity prices will move in future because returns have become a monetary phenomenon. The ERP is high because Treasury yields are unusually low at all maturities.
“With QE, things have changed. Asset prices are both the result of monetary action and a factor influencing it. The implied circularity is great when markets are doing well; but disastrous when they reverse. Notions of fair value, mean reversion and equilibrium price remain sidelined.”
Relative vs absolute valuation
The second salient point that emerged from our interviews was whether it is advisable to put too much faith in the ERP as a predictive tool, given that the traditional notion of ‘risk-free’ assets no longer holds, since there are no risk-free assets today (see case study).
For example, contrary to conventional wisdom, the US 10-year Treasury bond has outperformed almost all the risky assets in the past decade – often by an exceptional margin.
Far from being at the bottom of the returns pecking order, bonds have been at the top. In a balanced portfolio, equities used to be the risk engine and bonds the safety valve. The roles are now reversed. Bonds are an unsuitable benchmark for assessing the ERP.
As a result, the relative valuations between equities and bonds – as implied by the ERP – may be an inadequate guide to the future attractiveness or otherwise of equities. The implications are twofold.
First, in today’s climate, absolute – rather than relative – valuation is more relevant in equity investing. The deep-value approach stands a better chance of surviving the volatility while markets remain influenced more by politics than economics.
Second, the QE programmes have overly inflated the equity markets, lifting the good, the bad and the ugly. When the tide goes out – as it surely will one day – we shall see a huge dispersion in the returns of individual components of widely used indices such as the S&P500, FTSE 100 and MSCI AWI.
Current large inflows into indexed funds are making them inefficient by overly exposing them to momentum and concentration risks. Stock picking will be at a premium when valuations revert to their mean.
The ERP will remain volatile and unpredictable. Periodic spikes will be the norm as markets struggle to make sense of any new information. Investors would be wise not to read too much into it while it lacks a sensible anchor point for ‘risk-free’ assets.